Multi-strategy integrated Gorilla Troops Optimizer for solving global optimization and engineering design problems.

Area of Science:

• Artificial Intelligence
• Swarm Intelligence
• Optimization Algorithms

Background:

• The Artificial Gorilla Troops Optimizer (GTO) is effective for global exploration but suffers from premature convergence and local optima.
• Complex constraints and rugged search spaces exacerbate GTO's limitations.

Purpose of the Study:

• To introduce the Multi-Strategy Integrated Gorilla Troops Optimizer (MSIGTO) to overcome the limitations of the original GTO.
• To enhance the diversity, exploration, and convergence of optimization algorithms.

Main Methods:

• Integration of Latin Hypercube Sampling (LHS) for initial population diversity.
• Incorporation of Lévy Flight (LF) and Cauchy Inverse Cumulative Distribution Operator (CICDO) to improve exploration and convergence.
• Comparative analysis against 8 population-based optimization algorithms on CEC2017 and CEC2022 benchmark suites.

Main Results:

• MSIGTO achieved superior global exploration, convergence efficiency, and solution robustness.
• Demonstrated strong performance on high-dimensional (100D) and low-dimensional (20D) problems with Friedman mean ranks of 1.48 and 1.75, respectively.
• Validated effectiveness on four constrained real-world engineering problems.

Conclusions:

• MSIGTO significantly outperforms existing population-based algorithms.
• The algorithm shows broad potential for various engineering optimization applications.
• MSIGTO offers an effective solution for complex optimization challenges.